AntiretroviralTherapy for Control of the HIVassociated. Epidemic in Resource- Limited Settings

AntiretroviralTherapy for Control of the HIVassociated Tuberculosis Epidemic in Resource- Limited Settings Stephen D. Lawn, MD a,b, *,KatharinaKranzer,MD a,b, RobinWood, FCP MMed a KEYWORDS HIV Tuberculosis Antiretroviral Disease control Mortality HIV-ASSOCIATED TUBERCULOSIS EPIDEMIC Tuberculosis (TB) remains a major challenge to global public health and has proved particularly difficult to control in regions with high HIV prevalence. Of the estimated global burden of 9.3 million new TB cases in 2007, 1.37 million (14.8%) were associated with HIV. 1,2 TB remains a major cause of mortality, with an estimated 1.3 million deaths among individuals who are not infected with HIV and a further 0.5 million TB deaths among people who have HIV. The latter accounted for almost 25% of global AIDS-related mortality. HIV was a key factor underlying the approximately 1% annual increase in global TB incidence rates between 1990 and 2004. Sub-Saharan Africa has borne the brunt of this co-epidemic, accounting for 79% of cases of the global burden of HIV-associated TB in 2007. Here TB notification rates have increased three- to fivefold in many countries, especially in the south of the continent where HIV prevalence is highest. As a result, TB incidence rates in the region far exceed those observed in most other parts of the world (Fig. 1). Elsewhere, South and East Asia account for 11% of the global burden of HIV-associated TB and the combined epidemics of HIV and drug resistant TB are undermining TB control in Eastern Europe. 1 MILLENNIUM DEVELOPMENT GOALS FOR TUBERCULOSIS CONTROL The Millennium Development Goals (MDG) TB control targets are to halt and start to reverse the rising incidence of TB and to halve the 1990 prevalence and death rates by 2015. 3 Although the absolute number of TB cases occurring worldwide continued to increase in 2007, it was largely the result of global population growth. Global TB incidence rates are actually thought to have peaked in 2004 at 142 cases per 100,000 population and to have started to slowly decrease thereafter. 1,2 This recent downward trend in global TB incidence was preceded by a sustained decrease in S. D. Lawn and K. Kranzer are funded by the Wellcome Trust, London, UK. R.Wood is funded in part by the National Institutes of Health through a CIPRA grant 1U19AI53217 01 and RO1 grant (A1058736 01A1). a The Desmond Tutu HIV Centre, Institute of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, Anzio Road, Observatory 7925, Cape Town, South Africa b Department of Infectious and Tropical Diseases, Clinical Research Unit, London School of Hygiene and Tropical Medicine, London, UK * Corresponding author. The Desmond Tutu HIV Centre, Institute of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, Anzio Road, Observatory 7925, Cape Town, South Africa. E-mail address: stevelawn@yahoo.co.uk (S.D. Lawn). Clin Chest Med 30 (2009) 685 699 doi:10.1016/j.ccm.2009.08.010 0272-5231/09 ª 2009 Elsevier Inc. Open access under CC BY license. chestmed.theclinics.com

686 Lawn et al Fig. 1. Estimated TB incidence rates by country for 2006. (Reproduced from World Health Organization. Global Tuberculosis Control. Surveillance, planning, and financing. WHO/HTM/TB/2008.393. Geneva (Switzerland): World Health Organization; 2008; with permission.) global TB prevalence since 1990. 1 Rates of TB mortality (HIV-associated and non-hiv associated) are also estimated to have peaked around the year 2000 and have since decreased. 1 Despite these encouraging overall trends, however, the global MDG TB control target of halving the 1990 TB prevalence and death rates by 2015 will not be achieved with current rates of progress. 3 Targets are projected to be missed in the regions of sub-saharan Africa and Eastern Europe where the TB and HIV epidemics intersect. Concerted international public health action is needed. INTERNATIONAL PUBLIC HEALTH RESPONSE Since the World Health Organization (WHO) declaration in 1993 that TB was a global emergency, the directly observed treatment, shortcourse (DOTS) strategy has served as the key public health intervention that has been widely implemented to effect global TB control. 4 This strategy prioritizes the detection of patients who are sputum smear-positive presenting to health facilities with symptoms and aims to achieve high cure rates using short-course rifampicin-containing chemotherapy. Although this has been effective in most parts of the world, contributing to the sustained downward trend in global TB prevalence, it has nevertheless been comparatively ineffective in countries with high HIV prevalence. 1,2,5,6 Although the DOTS strategy remains the foundation for TB control programs in high HIV-prevalence settings, it is clear that additional interventions are needed. Under the leadership of the WHO and the Stop- TB Partnership, TB-HIV guidelines, 7 a TB-HIV strategic framework, 8 and an interim TB-HIV policy 9 were published between 2002 to 2004 to address the challenge of HIV-associated TB in severely affected countries. The aim of these interventions is to reduce the burden of TB in people who are HIV-infected through use of TB prevention strategies, such as isoniazid preventive therapy, intensified case finding, infection control, and antiretroviral therapy (ART). A further aim is to reduce the impact of HIV in patients who have TB through HIV testing and use of trimethoprim-sulfamethoxazole prophylaxis and ART. Great progress has been made over the past few years in HIV testing in patients who have TB 1 and in the scale-up of ART. 10 More than 3 million people in resource-limited settings were estimated to have started ART by the end of 2007 and 2 million of these were in sub-saharan Africa. 10 However, little is known about what impact this massive public health intervention will have on the HIV-associated TB epidemic or how ART might be used to best effect TB control. This article provides an in-depth review of these issues.

Antiretrovirals for HIV-Tuberculosis Control 687 TUBERCULOSIS INCIDENCE RATES IN ANTIRETROVIRAL TREATMENT COHORTS Data from 12 observational studies that included over 32,000 participants living in high-income countries (seven studies) and resource-limited countries (five studies) are shown in Table 1. In two of these studies, TB rates were simply compared before and after 1995 when tripledrug ART was introduced, whereas in the remaining studies TB rates were determined according to person-time of exposure to ART. All but two of these studies found a statistically significant reduction in TB risk during ART. In nine studies, the reductions in TB risk associated with use of antiretrovirals are expressed as hazard ratios adjusted for covariates, such as baseline patient characteristics and antiretroviral regimen. Triple-drug ART was associated with a risk reduction of more than 70% in the majority of these studies (n 5 6) with a range of 54% to 92% (see Table 1). This protective effect against TB was seen in countries with low-tb burden, such as the United States, 11 and South Africa, 12 which has the highest burden of HIV-associated TB in the world. 2 The benefits of ART are seen across a broad range of degrees of baseline immunodeficiency and clinical stage of disease, although the absolute reduction in TB rates is greatest in those with the most advanced HIV disease (Fig. 2). TB incidence rates in 14 ART cohorts are shown in Table 2. The observed rates are heterogeneous and are likely to vary according to the local TB incidence rates, the baseline degree of immunodeficiency, and the duration of ART. Several studies report rates stratified according to duration of ART and these are shown in Fig. 3. The figure illustrates the major differences between TB rates in various settings and the rapidity of the beneficial impact of ART. All of these studies show time-dependent reductions in TB rates, with most of the benefit occurring within the first 2 years of ART. 13 19 Although reductions in TB incidence in high- and low-tb burden settings are proportionately similar, 15 the greatest absolute reductions in TB risk of course occur in high-tb prevalence settings. Baseline risk factors for incident TB during ART in resource-limited settings included low-cd4 cell count, 12 15 advanced WHO stage of disease, 12,14 younger age, 14,15 low socioeconomic status, 12 past history of TB, 20 and male sex. 15 In highincome countries, baseline risk factors for TB included low-baseline CD4 count and HIV transmission category. 15,18 However, the risk of TB decreased as CD4 cell counts rose (Fig. 4). Thus, with increasing duration of treatment, the immunologic response to ART, rather than baseline patients characteristics, emerges as the dominant predictor of TB risk. 13 The strong relationship between TB rates and updated CD4 cell counts during ART is shown in Fig. 5. 19 Data are scarce concerning TB rates among those with optimum immune recovery. However, in a study in South Africa, subjects achieving CD4 cell counts of greater than 500 cells/ml retained a TB rate that was approximately twofold higher than that in subjects who were non- HIV infected living in the same community. 19 These data suggest that recovery of TB-specific immune function during ART may be incomplete even at high CD4 cell counts. ANTIRETROVIRALS AND TUBERCULOSIS INCIDENCE AT A COMMUNITY LEVEL Despite the extensive data that have arisen from observational cohort studies, empiric data regarding the impact of ART on TB incidence rates at the community level are lacking. This is a difficult issue to study. In-depth observational studies of sentinel communities with high TB and HIV prevalence during scale-up of ART are needed. Outcome measures should include the impact of ART scale-up on TB incidence and prevalence, TB transmission, and TB-associated mortality. However, trends in TB incidence and mortality over time will inevitably be confounded by other variables, such as the natural evolution of the HIV epidemic, changes over time in socio-economic factors, and the efficiency of TB control programs. Ecological studies comparing settings differing in the rapidity and coverage of ART scale-up may provide further insights. LIKELY LIMITED IMPACT OF ANTIRETROVIRALS AT COMMUNITY LEVEL Despite the major beneficial impact of ART on TB rates in observational cohorts, mathematical modeling studies suggest that the impact of ART on TB rates at a population level will be limited. 21 Various observations suggest that this conclusion is likely to be true 22 and are discussed in full in Box 1. Risk of Tuberculosis Before Starting Antiretroviral Treatment Following HIV seroconversion, risk of TB doubles and continues to increase as CD4 cell counts decline. 12,23 25 Patients, therefore, typically remain at a high risk of TB for many years before eligibility for ART. Further compounding this, HIV infection is often only diagnosed once patients

688 Lawn et al Table 1 Studies (n 5 12) reporting the impact of antiretroviral therapy on tuberculosis incidence rates in observational cohorts Study Setting N Study Period Study Design Studies comparing TB rates in cohorts before and after introduction of ART Brodt et al, 1997 57 Germany 1003 1992 1996 Cohort of homosexual men 1992 1996 Kirk et al, 2000 58 Europe 6,972 1994 1999 EuroSIDA multicenter cohort 1994 1999 Impact of ART ontb Incidence Rates No change in overall cohort incidence rates (range, 2.1 2.7 cases/100 PY) Overall rate in cohort decreased from 1.8 cases/100 PY to 0.3 cases/100 PY Studies comparing TB rates in patients receiving or not receiving ART Ledergerber et al, 1999 17 Switzerland 2410 1995 1997 Swiss HIV Cohort Study Rate 0.78 cases/100 PY pre- ART and 0.22 cases/100 PY during first 15 months ART Jones et al, 2000 11 United States 1992 1998 Multicenter cohort Adult/ Adolescent Spectrum of HIV Disease project Steep decreases in TB incidence rates Adjusted Hazards Ratio (95%CI) 0.2 (0.1 0.5) Girardi et al, 2000 59 Italy 1360 1995 1996 Multicenter cohort Not stated 0.08 (0.01 0.88) Santoro-Lopes et al, 2002 60 Brazil 255 1991 1998 Prospective cohort Not stated 0.2 (0.04 1.13) Badri et al, 2002 12 South Africa 1034 1992 2001 Rates compared in separate prospective observational cohorts receiving or not receiving ART Markedly lower TB rates across a broad spectrum of baseline CD4 counts and WHO stage 0.19 (0.09 0.38)

Golub et al, 2007 40 Brazil 11,026 2003 2005 Multicenter retrospective cohort Rates among those receiving and not receiving ART were 1.9 and 4.0 cases/100 PY, respectively 0.46 (0.33 0.63) Miranda et al, 2007 61 Brazil 463 1995 2001 Multicenter retrospective study Rates among those receiving and not receiving ART were 1.2 and 13.4 cases/100 PY, respectively 0.2 (0.1 0.6) Muga et al, 2007 62 Spain 2238 1980s 2004 Multicenter seroconverter cohort Marked reduction in rates after 1995 in all HIV transmission categories 0.31 (0.17 0.54) Moreno et al, 2008 63 Spain 4268 1997 2003 Multicenter hospital-based cohort Rates among those receiving and not receiving ART were 0.5 and 1.6 cases/100 PY, respectively 0.26 (0.16 0.40) Golub et al, 2009 64 South Africa 2778 2003 2007 Retrospective data from two study sites Abbreviation: PY, person-years. Rates among those receiving and not receiving ART were 4.6 and 7.1 cases/100 PY, respectively 0.36 (0.25 0.51) Antiretrovirals for HIV-Tuberculosis Control 689

690 Lawn et al TB incidence rate 30 25 20 15 10 5 0 3 or 4 1 or 2 <200 200-350 >350 WHO Stage present to medical services with advanced immunodeficiency and either have active TB or high risk of developing TB in the short term. 26 In addition to TB occurring before enrollment for ART, a substantial burden of prevalent TB may be diagnosed in ART services during screening before starting treatment. 13,16,27,28 The proportion detected may vary according to the intensity of investigation and availability of mycobacterial culture. In two studies from South Africa, 20% to 25% of adults referred for ART who did not already have a TB diagnosis were found to have sputum culture-positive TB, the majority of which was smear-negative. 27,28 The overall cumulative risk of TB before initiating ART may therefore be very high. In one ART service in Cape Town, this represented two thirds of adult patients. 13 Thus, ART is frequently implemented too late in the course of HIV progression to prevent much of the overall burden of HIVassociated TB. Incidence of Tuberculosis During Antiretroviral Therapy No ART ART Baseline CD4 Count Fig. 2. TB incidence (cases per 100 person-years) among patients who were HIV-infected in Cape Town, South Africa, who were or were not receiving antiretroviral therapy. Patients were stratified according to baseline CD4 cell count and WHO stage of disease. Overall, TB rates were approximately 80% lower among those receiving ART, which was observed across a broad spectrum of baseline immunodeficiency. (Data from Badri M, Wilson D, Wood R. Effect of highly active antiretroviral therapy on incidence of tuberculosis in South Africa: a cohort study. Lancet 2002;359(9323):2059 64). In addition to TB occurring pre-art, TB incidence rates during ART persist at rates substantially higher than background (see Table 2). During the initial months of treatment, a proportion of TB cases present as a result of the unmasking of subclinical TB during ART-induced restoration of TB-specific immune responses. 19,29,30 Through this mechanism, the risk of TB during ART may actually increase in the short term after starting ART. However, the size of this effect will be highly dependent on the efficiency with which patients are screened for clinical and subclinical TB before commencing ART. Unmasking TB was estimated to account for more than one third of the TB cases presenting during the first 4 months of ART in a treatment service in South Africa. 19 In the longer term, TB rates during ART appear to be largely dependent on changes in the CD4 cell count over time. With CD4 cell counts less than 200 cells/ml, TB rates exceeded 9.0 cases/ 100 person-years (PY) in the same study from South Africa. 19 Rates at higher CD4 counts were substantially lower, but nevertheless remained more than five-fold higher than rates in individuals who were non-hiv-infected living in the same community. 13,19 Persistence of high TB rates in ART cohorts is likely to reflect the fact that most patients spend prolonged periods at CD4 cell counts less than 500 cells/ml. In addition, immunologic studies also suggest that defects in functional TB-specific immune responses may persist despite ART, 31 leaving patients vulnerable to TB long-term. This may be compounded by high rates of nosocomial exposure to TB within overcrowded ART services where infection control procedures are often lacking. 32,33 Despite high early mortality risk, 34 long-term survival after the first year of ART is likely to be good in patients who are in treatment programs in sub-saharan Africa. In light of the persistence of high rates of TB during ART, the life-time cumulative risk of TB in survivors will inevitably be high. It is not yet clear whether this is reduced compared with the cumulative life-time risk in patients who are HIV-infected in the pre-art era. A further critical factor affecting the magnitude of the impact of ART on community TB rates is the coverage of ART. In mathematical modeling studies, high coverage with ART was needed to impact TB rates. 21 However, coverage in sub-saharan Africa was only approximately one third of those estimated to be in need at the end of 2007. 10 In most countries, rates of HIV-testing are low and only a small proportion of individuals who are HIV-infected know their HIV status. The magnitude of any impact of ART on TB transmission is unknown. It has been generally assumed that patients who are HIV-infected are not the key drivers of TB transmission in the community in view of their comparatively low infectivity. If true then ART is unlikely to have a major impact on community TB transmission. Conversely, however, patients receiving ART survive longer and remain at

Table 2 Studies (n514) reporting tuberculosis incidence rates during antiretroviral therapy Study Setting N High-income countries Girardi et al, 2005 18 Brinkhof et al, 2007 15 Germany, Switzerland, France, Netherland, UK, Canada, United States Europe, North America Median / Mean Followup (Months) Median Baseline CD4 Cell Count (Cells/mL) TB Cncidence, Cases/ 100 PY (Months of ART) 17,142 25.8 280 1.31 (0 3) 0.78 (4 6) 0.46 (7 12) 0.33 (13 24) 0.15 (25 36) 22,217 11.0 234 1.7 (0 3) 1.0 (4 6) 0.6 (7 12) Estimated National TB Incidence Rate (Per 100 Population) a 0.005 0.016 Moreno et al, 2008 63 Spain 4268 46.0 324 0.5 0.035 Resource-limited settings Badri et al, 2002 12 South Africa 1034 16.8 254 2.4 0.406 Santoro-Lopes et al, Brazil 284 22.0 8.4 0.071 2002 60 Lawn et al. 2005 14 South Africa 346 40.0 242 3.35 (0 12) 0.576 1.56 (13 24) 1.36 (25 36) 0.90 (37 48) 1.01 (49 60) Seyler et al, 2005 20 Côte d Ivoire 129 26.0 125 4.8 0.368 Lawn et al, 2006 13 South Africa 1002 0.9 96 23.0 (0 3) 10.7 (4 6) 7.0 (7 12) 3.7 (13 24) 0.898 (continued on next page) <0.015 Antiretrovirals for HIV-Tuberculosis Control 691

692 Table 2 (continued) Study Setting N Bonnet et al, 2006 65 Kenya Malawi Cameroon Thailand Cambodia 3151 3.7 6.7 11.1 3.7 7.3 Median / Mean Followup (Months) Median Baseline CD4 Cell Count (Cells/mL) 17.6 14.3 4.8 10.4 7.6 TB Cncidence, Cases/ 100 PY (Months of ART) Estimated National TB Incidence Rate (Per 100 Population) a Golub et al, 2007 40,64 Brazil 11,026 17.0 1.90 0.053 Miranda et al, 2007 61 Brazil 245 1.2 0.064 Brinkhof et al, 2007 15 Botswana, Brazil, 4540 9.6 107 10.7 (0 3) 0.055 0.852 Côte d Ivoire, India, Kenya, Nigeria, Malawi, Morocco, Senegal, South Africa, Thailand, Uganda 7.5 (4 6) 5.2 (7 12) Moore et al, 2007 16 Uganda 1044 17.0 127 3.9 (overall) 0.385 7.5 (0 6) 2.4 (7 12) 1.9 (13 18) Walters et al, 2008 66 South Africa 290 (pediatric) 6.4 0.898 0.419 0.416 0.194 0.142 Lawn et al Abbreviation: PY, person-years. a National TB incidence estimates at the midpoint of the study duration; data sourced from World Health Organization. Global tuberculosis control: epidemiology, strategy, financing. Geneva (Switzerland): World Health Organization; 2009. WHO/HTM/TB/2009.411.

Antiretrovirals for HIV-Tuberculosis Control 693 TB Incidence 25 20 15 10 5 0 0 12 24 36 48 60 Duration ART Fig. 3. TB incidence rates during ART. The graph shows data from studies included in (see Table 2) in which changing TB incidence rates were calculated according to increasing duration of ART. The two lowest curves present data from studies conducted in highincome countries. 15,18 The remaining four studies are from South Africa (diamonds 13 and inverted triangles 14 ), a range of resource-limited countries (circles 15 ), and Uganda (squares 16 ). substantial risk of TB that may be more likely to be sputum smear-positive as immune function improves. However, transmission risk among such patients remains undefined. Perhaps the most important issue relating to TB transmission is that occurring between individuals who are HIV-infected in health care settings. 33 This issue was exemplified by the outbreak of multidrug-resistant and extensively drug-resistant TB in a hospital-based ART service in rural South Africa from 2005 to 2006. 32 Although ART may serve as a potent preventive therapy for TB, ART services without adequate infection-control measures may nevertheless be sites associated with high risk of nosocomial TB transmission and outbreaks. ENHANCING THE IMPACT OF ANTIRETROVIRALS ON TUBERCULOSIS CONTROL Earlier Initiation of Antiretroviral Therapy It is likely that ART could be used much more effectively as a TB-prevention strategy (Box 2). ART should be initiated much earlier in the course of disease than is currently being done. In sub- Saharan Africa, the median CD4 cell count at baseline in ART programs is often in the range of 100 to 150 cells/ml 34 and many patients have already had TB before starting ART. Earlier HIV diagnosis would permit more timely initiation of ART, thereby enhancing its role as a preventive intervention. Consideration should be given to revision of ART eligibility guidelines in countries, such as South Africa, where treatment is restricted to patients who have WHO stage 4 disease or blood CD4 cell counts of less than 200 cells/ml. 35 A randomized controlled trial of early (CD4 200 350 cells/ml) versus late (CD4 <200 cells/ml) initiation of ART in Haiti was discontinued in 2009 following interim analysis. 36 In the delayed initiation group, the mortality was fourfold higher and the TB incidence rate was twofold higher than in the early initiation group, clearly demonstrating the benefits to be derived from earlier treatment. A mathematical modeling study explored the potential impact of a test and treat strategy TB incidence 35 30 25 20 15 10 5 450 400 350 300 250 200 150 100 50 Change in median CD4 count 0 0 1 2 3 Years of ART Fig. 4. Decreasing TB incidence rates (cases/100 person-years, white squares) and rising median CD4 cell counts (cells/ml, black diamonds) during the first 3 years of ART. These data are from a community-based ART cohort in a township in Cape Town, South Africa. (Data from Refs. 13,19,56 ). 0

694 Lawn et al TB incidence 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 Updated CD4 Count Fig. 5. Relationship between updated CD4 cell-count measurements made every 4 months during 4.5 years of ART in a treatment cohort in a township in Cape Town, South Africa. Observed rates are shown as diamonds together with 95% confidence intervals indicated by bars. A logarithmic trend line is overlaid (R 2 5 0.97). TB incidence rates are seen to fall substantially as CD4 cell counts increase during ART. (Data from Lawn SD, Myer L, Edwards D, et al. Short-term and long-term risk of tuberculosis associated with CD4 cell recovery during antiretroviral therapy in South Africa. AIDS 2009;23(13):1717 25.) whereby all patients who are diagnosed with HIV are immediately offered ART irrespective of the blood CD4 cell count. 38 Although interest has primarily focused on the potential of such a strategy for HIV prevention, this also has the potential to impact TB control. The rapidity of scale-up of ART may also be an important variable that affects not only the numbers of deaths averted by ART 39 but also the number of TB cases averted. Adjunctive Interventions Because TB rates do not decrease to background levels during long-term ART, use of concurrent adjunctive interventions is needed. Three key interventions are encompassed within the WHO 3I s strategy launched in April 2008. 37 This threepronged strategy includes intensified case finding, isoniazid preventive therapy, and infection control. Intensified case finding is needed in ART services as this designates patients as either having active TB (and in need of treatment) or being TB-free and potentially benefitting from isoniazid preventive therapy. Initiation of TB treatment in those who have disease also plays a critical role in reducing the risk of nosocomial TB transmission. Observational data have examined the potential role of isoniazid preventive therapy concurrently with ART in Brazil. Although isoniazid alone was not associated with significant reductions in TB risk compared with the nonintervention group, it was suggested that concurrent isoniazid preventive therapy during ART provided an additive benefit. 40 However, this requires confirmation by randomized controlled trials. Box 1 Reasons why scale-up of antiretroviral therapy is likely to have only limited impact on tuberculosis incidence rates at the community level Much HIV-associated TB occurs before initiation of ART High rates of TB persist during ART Prolonged survival of patients on ART is associated with a high life-time cumulative risk of TB Limited coverage of ART in the community and limited compliance with treatment Impact of ART on TB transmission in the community is suspected to be low Box 2 Potential ways to enhance the impact of antiretroviral therapy on tuberculosis prevention at a community level Diagnose HIV earlier and initiate ART at higher CD4 cell counts Increase coverage of ART in the community Increase rapidity of scale-up of ART Use of adjunctive interventions with ART, such as those included within the WHO 3I s strategy (intensified case finding, isoniazid preventive therapy and infection control) 37

Table 3 Observational cohort studies (n58) showing the impact of antiretroviral therapy on mortality among patients who have HIV-associated tuberculosis Study Country Study Design Outcome Dheda et al, 2004 67 United Kingdom Retrospective study of patients who had HIV-TB (n 5 99) treated in pre-art era and in ART era Manosuthi et al, 2006 68 Thailand Retrospective cohort study (n 5 1003) comparing mortality in a historic natural- history cohort with rates in an ART cohort Akksilp et al, 2007 69 Thailand Prospective cohort (n 5 329) comparing patients receiving and not receiving ART Zachariah et al, 2007 43 Malawi Retrospective observational cohort in which a proportion of patients started ART during the continuation phase of TB treatment (n 5 658) Nahid et al, 2007 49 United States Retrospective observational cohort (n 5 264) 1990 2001 spanning pre-art and ART era Haar et al, 2007 42 Netherlands Retrospective observational study of national data 1993 2001 spanning pre-art and ART era Varma et al, 2009 70 Thailand Prospective multicenter observational study (n 5 667) comparing patients receiving and not receiving ART Velasco et al, 2009 71 Spain Retrospective observational cohort 1987 2004 (n 5 313) comparing patients receiving and not receiving ART Adjusted hazards of death or new AIDS-defining illness was 0.34 (95%CI, 0.18 0.63) during ART era The adjusted hazards of death associated with use of ART was 0.05 (95%CI, 0.02 0.12). Adjusted hazards of death was 0.2 (95%CI, 0.1 0.4) No difference in mortality between patients who chose or did not choose to receive ART, but potential allocation bias according to degree of immunodeficiency and most deaths occurred pre- ART during intensive phase Use of ART protected against mortality compared with patients who did not receive ART (hazard ratio 0.36, 95% CI 0.14 0.91) Compared with 1993 1995, adjusted odds of death during 1999 2001 was 0.46 (95%CI, 0.24 0.89), whereas no such change was observed among patients who had TB and were not infected with HIV Adjusted hazards of death among those who received ART was 0.16 (95%CI, 0.07 0.36) Compared with no ART, initiation of ART within the first 2 months of TB treatment was associated with an adjusted hazards of death of 0..37 (95%CI, 0.17 0.66) Antiretrovirals for HIV-Tuberculosis Control 695

696 Lawn et al IMPACT OF ANTIRETROVIRALS ON MORTALITY RISK TB case fatality rates (the proportion of patients dying while receiving antituberculosis treatment) in Africa are 16% to 35% among patients who are infected with HIV and not receiving ART and 4% to 9% among individuals who were not infected with HIV. 41 The authors identified observational studies (n 5 8) that provided information on the impact of ART on survival of patients who have HIV-associated TB (Table 3). Of these, half were from high-income countries and half were from resource-limited settings. One study using national data from The Netherlands found a 54% reduction in risk of death of patients who had HIV-associated TB treated in the ART era compared with in the pre-art era. 42 Of seven further studies, six found that use of ART was associated with significant reductions in mortality risk of 64% to 95% in adjusted analyses (see Table 3). The remaining study from Malawi found no such association but analyses were not adjusted for CD4 cell counts and ART was not commenced during the intensive phase of TB treatment when most deaths occurred. 43 A key issue that remains to be resolved regarding the clinical management of patients who have HIV-associated TB is the optimal time to start ART during TB treatment. Early mortality rates are extremely high among patients waiting to start ART in resource-limited settings. 44,45 However, early initiation of ART in those who have HIV-associated TB increases the risk of TB immune reconstitution disease. 46 This increase is associated with mortality risk in a small proportion of patients 46,47 but may be greater in those who have neurologic involvement. 48 Overall, the risk/ benefit ratio is likely to favor initiation of ART in the initial weeks of TB treatment, but the outcome of randomized controlled trials are awaited. ANTIRETROVIRALS AND TUBERCULOSIS TREATMENT OUTCOMES Data concerning the impact of ART on TB treatment outcomes other than death are scarce. A large retrospective observational cohort study in San Francisco reported that use of ART was associated with significant shortening of the time to smear and culture conversion. 49 A systematic review of 32 studies 50 and recently published data from Rio de Janeiro, South African gold mines, and San Francisco 49,51,52 confirm that patients who have HIV-associated TB are at increased risk of recurrent disease, especially those who have low CD4 cell counts. Various strategies can be used to address high recurrence rates, including prolongation of the duration of TB therapy 49,53 and use of secondary isoniazid prophylaxis. 54,55 However, more recent data from a study in Brazil have now shown that recurrence rates were halved in patients who had TB who received ART. 51 It remains to be determined whether combining ART and isoniazid has an additive effect. SUMMARY The HIV-associated TB epidemic is undermining progress toward TB control and additional interventions must be used in combination with the DOTS strategy in resource-limited settings. Observational cohort studies in a wide range of settings have demonstrated that the use of ART is associated with a 54% to 92% reduction in TB incidence rates and a halving of the risk of TB recurrence. ART is also a potent means of improving the survival of those who have HIVassociated TB, reducing mortality rates by 64% to 95%. Thus, ART confers huge benefits to individual patients. However, limited data are available concerning the effects of ART on TB rates at a population level. Mathematical modeling and a range of empiric observations suggest that this may be limited, largely because of the high burden of TB that occurs before ART initiation and the persistence of TB rates several-fold higher than background rates during long-term ART. Thus, ART is likely to have limited impact on lifetime cumulative risk of TB. To enhance the impact on community TB control, it is likely that ART will need to be implemented with high-population coverage and at much higher CD4 cell counts than is currently the case. Adjunctive interventions, such as those included in the 3I s policy, may help reduce rates further. Despite the likely limited impact on TB rates at a community level, ART nevertheless transforms the prognosis of patients who have HIV-associated TB. By reducing mortality rates, ongoing ART scale-up may accelerate progress toward the MDG TB control target of halving the 1990 mortality rate by 2015. REFERENCES 1. World Health Organization. Global tuberculosis control: epidemiology, strategy, financing. Geneva (Switzerland): World Health Organization; 2009. WHO/HTM/TB/2009.411. 2. Lawn SD, Churchyard G. Epidemiology of HIV-associated tuberculosis. Curr Opin HIV AIDS 2009;4:325 33.

Antiretrovirals for HIV-Tuberculosis Control 697 3. United Nations. The millennium development goals report 2008. Available at: http://www.un.org/ millenniumgoals. Accessed June 8, 2009. 4. Frieden TR, Munsiff SS. The DOTS strategy for controlling the global tuberculosis epidemic. Clin Chest Med 2005;26(2):197 205. 5. De Cock KM, Chaisson RE. Will DOTS do it? A reappraisal of tuberculosis control in countries with high rates of HIV infection. Int J Tuberc Lung Dis 1999; 3(6):457 65. 6. Lawn SD, Bekker LG, Middelkoop K, et al. Impact of HIV infection on the epidemiology of tuberculosis in a peri-urban community in South Africa: the need for age-specific interventions. Clin Infect Dis 2006; 42(7):1040 7. 7. World Health Organization. Guidelines for implementing collaborative TB and HIV programme activities. Available at: http://www.who.int/tb/publications/ 2003/en/index1.html http://www.who.int/hiv/pub/ prev_care/pub31/en/. Accessed February 20, 2008. 8. World Health Organization. Strategic framework to decrease the burden of TB/HIV. Available at: http:// www.who.int/tb/publications/who_cds_tb_2002_296/ en/index.html. Accessed August 1, 2009. 9. World Health Organization. Interim policy on collaborative TB/HIV activities 2004. Available at: http:// whqlibdoc.who.int/hq/2004/who_htm_tb_2004. 330_eng.pdf. Accessed August 1, 2009. 10. World Health Organization. Towards universal access. Scaling up priority HIV/AIDS interventions in the health sector. Progress report 2008. Available at: http://www.who.int/entity/hiv/pub/towards_universal_ access_report_2008.pdf. Accessed August 1, 2009. 11. Jones JL, Hanson DL, Dworkin MS, et al. HIV-associated tuberculosis in the era of highly active antiretroviral therapy. The Adult/Adolescent Spectrum of HIV Disease Group. Int J Tuberc Lung Dis 2000; 4(11):1026 31. 12. Badri M, Wilson D, Wood R. Effect of highly active antiretroviral therapy on incidence of tuberculosis in South Africa: a cohort study. Lancet 2002; 359(9323):2059 64. 13. Lawn SD, Myer L, Bekker LG, et al. Burden of tuberculosis in an antiretroviral treatment programme in sub-saharan Africa: impact on treatment outcomes and implications for tuberculosis control. AIDS 2006;20(12):1605 12. 14. Lawn SD, Badri M, Wood R. Tuberculosis among HIV-infected patients receiving HAART: long term incidence and risk factors in a South African cohort. AIDS 2005;19(18):2109 16. 15. Brinkhof MW, Egger M, Boulle A, et al. Tuberculosis after initiation of antiretroviral therapy in low-income and high-income countries. Clin Infect Dis 2007; 45(11):1518 21. 16. Moore D, Liechty C, Ekwaru P, et al. Prevalence, incidence and mortality associated with tuberculosis in HIV-infected patients initiating antiretroviral therapy in rural Uganda. AIDS 2007;21(6):713 9. 17. Ledergerber B, Egger M, Erard V, et al. AIDS-related opportunistic illnesses occurring after initiation of potent antiretroviral therapy: the Swiss HIV Cohort Study. JAMA 1999;282(23):2220 6. 18. Girardi E, Sabin CA, d Arminio MA, et al. Incidence of tuberculosis among HIV-infected patients receiving highly active antiretroviral therapy in Europe and North America. Clin Infect Dis 2005; 41(12):1772 82. 19. Lawn SD, Myer L, Edwards D, et al. Short-term and long-term risk of tuberculosis associated with CD4 cell recovery during antiretroviral therapy in South Africa. AIDS 2009;23(13):1717 25. 20. Seyler C, Toure S, Messou E, et al. Risk factors for active tuberculosis after antiretroviral treatment initiation in Abidjan. Am J Respir Crit Care Med 2005; 172(1):123 7. 21. Williams BG, Dye C. Antiretroviral drugs for tuberculosis control in the era of HIV/AIDS. Science 2003; 301(5639):1535 7. 22. Lawn SD, Wood R. Tuberculosis control in South Africa will HAART help? S Afr Med J 2006;96(6): 502 4. 23. Sonnenberg P, Glynn JR, Fielding K, et al. How soon after infection with HIV does the risk of tuberculosis start to increase? A retrospective cohort study in South African gold miners. J Infect Dis 2005; 191(2):150 8. 24. Glynn JR, Murray J, Bester A, et al. Effects of duration of HIV infection and secondary tuberculosis transmission on tuberculosis incidence in the South African gold mines. AIDS 2008;22(14):1859 67. 25. Holmes CB, Wood R, Badri M, et al. CD4 decline and incidence of opportunistic infections in Cape Town, South Africa: implications for prophylaxis and treatment. J Acquir Immune Defic Syndr 2006; 42(4):464 9. 26. Lawn SD, Wood R. How can earlier entry of patients into antiretroviral programs in low-income countries be promoted? Clin Infect Dis 2006;42(3):431 2. 27. Lawn SD, Edwards SD, Kranzer K, et al. Urine lipoarabinomannan assay for tuberculosis screening prior to ART: diagnostic yield and association with immune reconstitution disease. AIDS 2009;23(14): 1875 80. 28. Bassett I, Chetty S, Wang B, et al. Intensive TB screening for HIV-infected patients ready to start ART in Durban, South Africa: limitations of WHO guidelines [Abstract #779]. Program and Abstracts of the 16th Conference on Retroviruses and Opportunistic Infections. Montreal (Canada), February, 2009. 29. Breen RA, Smith CJ, Cropley I, et al. Does immune reconstitution syndrome promote active tuberculosis in patients receiving highly active antiretroviral therapy? AIDS 2005;19(11):1201 6.

698 Lawn et al 30. Lawn SD, Wilkinson RJ, Lipman MC, et al. Immune reconstitution and unmasking of tuberculosis during antiretroviral therapy. Am J Respir Crit Care Med 2008;177(7):680 5. 31. Lawn SD, Bekker LG, Wood R. How effectively does HAART restore immune responses to Mycobacterium tuberculosis? Implications for tuberculosis control. AIDS 2005;19(11):1113 24. 32. Gandhi NR, Moll A, Sturm AW, et al. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet 2006; 368(9547):1575 80. 33. Bock NN, Jensen PA, Miller B, et al. Tuberculosis infection control in resource-limited settings in the era of expanding HIV care and treatment. J Infect Dis 2007;196(Suppl 1):S108 13. 34. Lawn SD, Harries AD, Anglaret X, et al. Early mortality among adults accessing antiretroviral treatment programmes in sub-saharan Africa. AIDS 2008;22(15):1897 908. 35. Lawn SD, Wood R. National adult antiretroviral therapy guidelines in South Africa: concordance with 2003 WHO guidelines? AIDS 2007;21(1):121 2. 36. NIAID. The CIPRA HT 001 clinical trial. Available at: http://www3.niaid.nih.gov/news/qa/cipra_ht01_ qa.htm. Accessed June 10, 2009. 37. World Health Organization. WHO three I s meeting. Report of a joint WHO HIV/AIDS and TB department meeting 2008. Available at: http://www.who.int/hiv/ pub/meetingreports/who_3is_meeting_report.pdf. Accessed August 1, 2009. 38. Granich RM, Gilks CF, Dye C, et al. Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model. Lancet 2009;373(9657): 48 57. 39. Walensky RP, Wood R, Weinstein MC, et al. Scaling up antiretroviral therapy in South Africa: the impact of speed on survival. J Infect Dis 2008;197(9): 1324 32. 40. Golub JE, Saraceni V, Cavalcante SC, et al. The impact of antiretroviral therapy and isoniazid preventive therapy on tuberculosis incidence in HIV-infected patients in Rio de Janeiro, Brazil. AIDS 2007;21(11):1441 8. 41. Mukadi YD, Maher D, Harries A. Tuberculosis case fatality rates in high HIV prevalence populations in sub-saharan Africa. AIDS 2001;15(2):143 52. 42. Haar CH, Cobelens FG, Kalisvaart NA, et al. HIVrelated mortality among tuberculosis patients in The Netherlands, 1993 2001. Int J Tuberc Lung Dis 2007;11(9):1038 41. 43. Zachariah R, Fitzgerald M, Massaquoi M, et al. Does antiretroviral treatment reduce case fatality among HIV-positive patients with tuberculosis in Malawi? Int J Tuberc Lung Dis 2007;11(8): 848 53. 44. Lawn SD, Myer L, Orrell C, et al. Early mortality among adults accessing a community-based antiretroviral service in South Africa: implications for programme design. AIDS 2005;19(18):2141 8. 45. Fairall LR, Bachmann MO, Louwagie GM, et al. Effectiveness of antiretroviral treatment in a South african program: a cohort study. Arch Intern Med 2008;168(1):86 93. 46. Lawn SD, Myer L, Bekker LG, et al. Tuberculosis-associated immune reconstitution disease: incidence, risk factors and impact in an antiretroviral treatment service in South Africa. AIDS 2007;21(3):335 41. 47. Manosuthi W, Kiertiburanakul S, Phoorisri T, et al. Immune reconstitution inflammatory syndrome of tuberculosis among HIV-infected patients receiving antituberculous and antiretroviral therapy. J Infect 2006;53(6):357 63. 48. Pepper DJ, Marais S, Maartens G, et al. Neurologic manifestations of paradoxical tuberculosis-associated immune reconstitution inflammatory syndrome: a case series. Clin Infect Dis 2009;48(11):e96 107. 49. Nahid P, Gonzalez LC, Rudoy I, et al. Treatment outcomes of patients with HIV and tuberculosis. Am J Respir Crit Care Med 2007;175(11):1199 206. 50. Panjabi R, Comstock GW, Golub JE. Recurrent tuberculosis and its risk factors: adequately treated patients are still at high risk. Int J Tuberc Lung Dis 2007;11(8):828 37. 51. Golub JE, Durovni B, King BS, et al. Recurrent tuberculosis in HIV-infected patients in Rio de Janeiro, Brazil. AIDS 2008;22(18):2527 33. 52. Charalambous S, Grant AD, Moloi V, et al. Contribution of reinfection to recurrent tuberculosis in South African gold miners. Int J Tuberc Lung Dis 2008; 12(8):942 8. 53. Perriens JH, St Louis ME, Mukadi YB, et al. Pulmonary tuberculosis in HIV-infected patients in Zaire. A controlled trial of treatment for either 6 or 12 months. N Engl J Med 1995;332(12):779 84. 54. Fitzgerald DW, Desvarieux M, Severe P, et al. Effect of post-treatment isoniazid on prevention of recurrent tuberculosis in HIV-1-infected individuals: a randomised trial. Lancet 2000;356(9240):1470 4. 55. Churchyard GJ, Fielding K, Charalambous S, et al. Efficacy of secondary isoniazid preventive therapy among HIV-infected Southern Africans: time to change policy? AIDS 2003;17(14):2063 70. 56. Lawn SD, Little F, Bekker LG, et al. Changing mortality risk associated with CD4 cell response to antiretroviral therapy in South Africa. AIDS 2008;23:335 42. 57. Brodt HR, Kamps BS, Gute P, et al. Changing incidence of AIDS-defining illnesses in the era of antiretroviral combination therapy. AIDS 1997; 11(14):1731 8.

Antiretrovirals for HIV-Tuberculosis Control 699 58. Kirk O, Gatell JM, Mocroft A, et al. EuroSIDA Study Group JD. Infections with Mycobacterium tuberculosis and Mycobacterium avium among HIV-infected patients after the introduction of highly active antiretroviral therapy. Am J Respir Crit Care Med 2000; 162(3 Pt 1):865 72. 59. Girardi E, Antonucci G, Vanacore P, et al. Impact of combination antiretroviral therapy on the risk of tuberculosis among persons with HIV infection. AIDS 2000;14(13):1985 91. 60. Santoro-Lopes G, de Pinho AM, Harrison LH, et al. Reduced risk of tuberculosis among Brazilian patients with advanced human immunodeficiency virus infection treated with highly active antiretroviral therapy. Clin Infect Dis 2002;34(4):543 6. 61. Miranda A, Morgan M, Jamal L, et al. Impact of antiretroviral therapy on the incidence of tuberculosis: the Brazilian experience, 1995 2001. PLoS One 2007;2(9):e826. 62. Muga R, Ferreros I, Langohr K, et al. Changes in the incidence of tuberculosis in a cohort of HIV-seroconverters before and after the introduction of HAART. AIDS 2007;21(18):2521 7. 63. Moreno S, Jarrin I, Iribarren JA, et al. Incidence and risk factors for tuberculosis in HIV-positive subjects by HAART status. Int J Tuberc Lung Dis 2008; 12(12):1393 400. 64. Golub JE, Pronyk P, Mohapi L, et al. Isoniazid preventive therapy, HAART and tuberculosis risk in HIV-infected adults in South Africa: a prospective cohort. AIDS 2009;23(5):631 6. 65. Bonnet MM, Pinoges LL, Varaine FF, et al. Tuberculosis after HAART initiation in HIV-positive patients from five countries with a high tuberculosis burden. AIDS 2006;20(9):1275 9. 66. Walters E, Cotton MF, Rabie H, et al. Clinical presentation and outcome of tuberculosis in human immunodeficiency virus infected children on anti-retroviral therapy. BMC Pediatr 2008;8:1. 67. Dheda K, Lampe FC, Johnson MA, et al. Outcome of HIV-associated tuberculosis in the era of highly active antiretroviral therapy. J Infect Dis 2004; 190(9):1670 6. 68. Manosuthi W, Chottanapand S, Thongyen S, et al. Survival rate and risk factors of mortality among HIV/tuberculosis-coinfected patients with and without antiretroviral therapy. J Acquir Immune Defic Syndr 2006;43(1):42 6. 69. Akksilp S, Karnkawinpong O, Wattanaamornkiat W, et al. Antiretroviral therapy during tuberculosis treatment and marked reduction in death rate of HIVinfected patients, Thailand. Emerg Infect Dis 2007; 13(7):1001 7. 70. Varma JK, Nateniyom S, Akksilp S, et al. HIV care and treatment factors associated with improved survival during TB treatment in Thailand: an observational study. BMC Infect Dis 2009;9:42. 71. Velasco M, Castilla V, Sanz J, et al. Effect of simultaneous use of highly active antiretroviral therapy on survival of HIV patients with tuberculosis. J Acquir Immune Defic Syndr 2009;50(2):148 52.